|Pinter jr, Paul|
|Hunsaker, Douglas - Doug|
Submitted to: Remote Sensing of Environment
Publication Type: Peer reviewed journal
Publication Acceptance Date: 5/26/2005
Publication Date: 8/29/2005
Citation: Fitzgerald, G.J., Pinter Jr, P.J., Hunsaker, D.J., Clarke, T.R. 2005. Multiple shadow fractions in spectral mixture analysis of a cotton canopy. Remote Sensing of Environment. 97:526-539. Interpretive Summary: Shadows are being used more frequently to estimate plant canopy biophysical characteristics. Typically, a zero value is assumed or a threshold value is derived from histogram analysis of imagery to determine the shadow endmember (EM). Here, two distinct shadow EMs were measured in situ for use in spectral mixture analysis of a cotton canopy on five dates in 2003. The four EMs used in the analysis were: sunlit green leaf, sunlit dry soil, shadowed leaf, shadowed dry soil. This 4-EM model was compared to a 3-EM model where a zero-value shade EM was used for unmixing with the two sunlit EMs. Multiple endmember spectral mixture analysis (MESMA) was used to allow EM composition to vary across each scene. The analysis and EMs were applied to fine-scale hyperspectral image data collected in the wavelength range, 440 to 810 nm. Ground data collected included percent cover, height, SPAD (a measure of leaf greenness), and chlorophyll a. The normalized difference vegetation index (NDVI) was also compared to the unmixing results. Regression analysis showed that NDVI was slightly superior to both mixture models for estimation of percent cover (r2 = 0.96, RMSE = 6.1). Different combinations of EMs from the 4-EM model were best at estimating height (r2 = 0.78, RMSE = 0.073 m), SPAD (r2 = 0.58, RMSE = 3.3), and chlorophyll a (r2 = 0.46, RMSE = 7.0 µg/cm2). The 3-EM model and NDVI performed poorly when estimating SPAD and chlorophyll a. Inclusion of two distinct shadow EM in the model improved relationships to crop biophysical parameters and was better than assuming one, zero-value shade EM. Since MESMA operates at the pixel level and allows variable EM assignment to each pixel, mapping the spatial variability of shadows and other variables of interest is possible, providing a powerful input to canopy and ecosystem models as well as precision farming.
Technical Abstract: Remotely-sensed images of agricultural fields can provide an array of information about the components in a field, including plant health, soil, stressed plants, and shadows. Spectral mixture analysis is an advanced image processing method that can estimate the type and amount of each component within each pixel of an image. The method is generally used to analyze hyperspectral imagery, which is composed of dozens to hundreds of wavebands from which spectra can be extracted. Spectral mixture analysis compares the actual pixel spectra with a reference library of spectra of 'pure' components and determines the fraction of each component contained in a pixel. One component, shadows, can either confound analysis of imagery or provide new information to improve estimation of ground features. In this study, two different types of shadows were identified to improve the ability of hyperspectral imagery to estimate the four crop factors, percent cover, crop height, SPAD (a measure of leaf greenness), and leaf chlorophyll. This approach has the potential to map various canopy factors simultaneously for use as inputs to crop and ecological models as well as provide spatially variable information for variable rate application of inputs in precision farming.